Vocal learning is a skill that is critical for the development of human language. Experimental study of vocal learning has been hampered by the paucity of animal models of vocal learning. Cetaceans are the only mammals other than man demonstrated to be skilled at vocal mimicry. Dolphins can be trained to mimic man made sounds and can learn to associate these sounds with objects or actions in artificial communication systems (Herman et al. 1984, Richards et al. 1984). The proposed research will test how dolphins use these skills in developing their natural systems of communication. Bottlenosed dolphins develop individually distinctive "signature whistles" (Caldwell and Caldwell 1979), and they apparently learn to imitate the signature whistles of members of their social groups (Tyack 1986). The research proposed here investigates the role of vocal learning and mimicry in the ontogeny of whistle development in young dolphins. A passive acoustic localization technique will be used to determine the whistle repertoires of young dolphins and other members of their natal groups several times/year for several years. The specific aims of this study are to answer the following questions in several groups of captive and wild dolphins. Experiment 1. Identify the different stages in the ontogeny of sterotyped whistles. When do young dolphins first learn to imitate the signature whistle of their mother or of other dolphins? When do young dolphins first develop their own signature whistle? In what social contexts do young dolphins produce their signature, initate the signature of their mother, or of other dolphins? Do other dolphins show a specific response when calves imitate their signatures? Do calves respond specifically when others imitate their signature? Experiment 2. What is the role of vocal learning and of early acoustic environment on the development of a young dolphin's own signature whistle and any other stereotyped whistles it produces. Inverse dialect hypothesis: Do young dolphins develop signature whistles that are particularly distinctive with respect to those of their natal group? Signature mimicry hypothesis: Do dolphins learn to mimic signature whistles of other dolphins through exposure to them? Both of these hypotheses will be tested by comparing the stereotyped whistles produced by young dolphins with the signature whistles from their own and from different groups. These studies promise to provide a comparative model in a non- human mammal for the roles of vocal learning and mimicry in the development of a natural communcation system.